Juntao Yu

Post-Doctoral Researcher in Natural Language Processing
School of Electronic Engineering and Computer Science
Queen Mary University of London
London E1 4FZ
Office: CS413
Email: juntao.yu AT qmul.ac.uk



I am a post-doctoral researcher at the Queen Mary University of London, working with Professor Massimo Poesio on his five-year DALI project (Disagreements and Language Interpretation, ERC-2015-AdG). Before that, I was a PhD student at the University of Birmingham, working on out-of-domain dependency parsing. I am supervised by Dr Bernd Bohnet, Professor John Barnden and Dr Mark Lee. I received my Master degree from the University of Glasgow in 2013. Before that, I worked at Digital China (Shanghai) Ltd. as a software engineer after I received my undergraduate degree in 2009 from Donghua University (Shanghai, China).

My research interests include anaphora resolution, dependency parsing, deep learning, domain adaptation, semi-supervised methods, unsupervised methods and language model.


My Google Scholar page.

The latest version of my CV.


Publications

  • Juntao Yu and Bernd Bohnet. 2017. Dependency Language Models for Transition-based Dependency Parsing. In Proceeding of the 15th International Conference on Parsing Technologies (IWPT), pages 11-17, Pisa, Italy, September. Association for Computational Linguistics [pdf] [bib]

  • Juntao Yu and Bernd Bohnet. 2015. Exploring Confidence-based Self-training for Multilingual Dependency Parsing in an Under-Resourced Language Scenario. In Proceeding of the Third International Conference on Dependency Linguistics (Depling), pages 350-358, Uppsala, Sweden, August. Uppsala University. [pdf] [bib]

  • Juntao Yu, Mohab Elkaref, and Bernd Bohnet. 2015. Domain Adaptation for Dependency Parsing via Self-training. In Proceeding of the 14th International Conference on Parsing Technologies (IWPT), pages 1-10, Bilbao, Spain, July. Association for Computational Linguistics. [pdf] [bib]

  • Viktor Pekar, Juntao Yu, Mohab Elkaref, and Bernd Bohnet. 2014. Exploring Options for Fast Domain Adaptation of Dependency Parsers. In Proceedings of the First Joint Workshop on Statistical Parsing of Morphologically Rich Languages and Syntactic Analysis of Non-Canonical Languages (SPMRL-SANCL), pages 54-65, Dublin, Ireland, August. Dublin City University. [pdf] [bib]

Teaching

University of Birmingham (Teaching Assistant)
  • Software Workshop 1 (Autumn 2015 - Autumn 2016)
  • Intro to AI (Autumn 2016)
  • Team Project (Spring 2015)
  • Natural Language Processing 1 (Spring 2015)

Events

Sep 2017: Attending IWPT to present our paper on dependency language models
Sep 2015: Attending Google NLP Summit.
Aug 2015: Attending Depling to present a paper on multilingual self-training.
Jul 2015: Attending IWPT to present our paper on self-training for domain adaptation.
Aug 2014: Attending Coling to present our workshop paper.


Last Update: September 2017